Is Google's Comment Filtering Tool 'Vanishing' Legitimate Comments? (vortex.com)

Slashdot reader Lauren Weinstein writes: Google has announced (with considerable fanfare) public access to their new "Perspective" comment filtering system API, which uses Google's machine learning/AI system to determine which comments on a site shouldn't be displayed due to perceived high spam/toxicity scores. It's a fascinating effort. And if you run a website that supports comments, I urge you not to put this Google service into production, at least for now.

The bottom line is that I view Google's spam detection systems as currently too prone to false positives -- thereby enabling a form of algorithm-driven "censorship" (for lack of a better word in this specific context) -- especially by "lazy" sites that might accept Google's determinations of comment scoring as gospel... as someone who deals with significant numbers of comments filtered by Google every day -- I have nearly 400K followers on Google Plus -- I can tell you with considerable confidence that the problem isn't "spam" comments that are being missed, it's completely legitimate non-spam, non-toxic comments that are inappropriately marked as spam and hidden by Google.
Lauren is also collecting noteworthy experiences for a white paper about "the perceived overall state of Google (and its parent corporation Alphabet, Inc.)" to better understand how internet companies are now impacting our lives in unanticipated ways. He's inviting people to share their recent experiences with "specific Google services (including everything from Search to Gmail to YouTube and beyond), accounts, privacy, security, interactions, legal or copyright issues -- essentially anything positive, negative, or neutral that you are free to impart to me, that you believe might be of interest."

Comments

Really?

[Webs 101]

There are 400,000 users of Google+?

automated

[Lehk228]

so now we are automating social justice? what are the college students going to do with their humanities degrees?

Re:automated

[AmiMoJo]

Be careful with this one. There are exactly zero examples in TFA, no evidence of what is being claimed. It's just an anecdote at this point.

Any kind of filter will never be 100% perfect. That's just the nature of filters. Occasionally the odd bit of spam will get through, the odd legit message will be marked as spam. That doesn't mean they are not useful. We should wait for actual data before making a judgement here.

Ultimately it's up to the individual. If you ant full uncensored speech then turn off your spam filter and handle the torrent manually. If you have better thing to do with your life and your channel is just Pokemon Go videos anyway, something like this could be useful.

NaturalNews violated Google TOS, not censored.

[Anonymous Coward]

Google Webmaster Tends Analyst John Mueller:

"Hi! I work with the Google Search team. We’re seeing a bit of confusion & incorrect stories circulating about what’s happening here, so just to be super clear — Natural News is using a sneaky mobile redirect, which is prohibited by our webmaster guidelines (there’s a bit about this kind of issue at https://webmasters.googleblog.com/2015/10/detect-and-get-rid-of-unwanted-sneaky.html). These redirects aren’t always easy to reproduce, they’re sometimes in widgets or served by ad networks, and can target specific devices, browsers, or user locations. When we last checked, there was one on blogs .naturalnews. com/bentonite-clay-a-natural-medicine-cabinet-must-have/. As soon as this is cleaned up, the site can submit a reconsideration request through Search Console, and once that’s reviewed things will return to normal. No action has been taken based on the editorial content of this site."

https://productforums.google.com/forum/#!topic/webmasters/3BNKoRXA49g/discussion

Slashdot

[darkain]

Slashdot's Meta-Moderation is by no means perfect, but it is a hell of a lot better than 99% of the web sites out there, especially anything that has automated moderation. Don't feel like dealing with assholes? Then don't browse at -1. Odds are someone else with karma has already come along and moderated the assholes into oblivion.

Re:Slashdot

[Anonymous Coward]

No, they moderated someone they disagree with into oblivion. Want to read real comments (including all the garbage) - browse at -1 to see what people are actually saying, not what the current no-life karma whores deem accepted in the current agenda/groupthink.

I only post AC on slashdot. Why? Because I cannot be banned by other users.I registered to comment once. I made several simple non-spam posts of my opinions on some windows tech topic, and was promptly banned from posting - by other users! I guess they did not want people to see what I had to say.

Slashdot's "meta-moderation" system is cowardly censorship hiding behind sockpuppet accounts, Want to foster real discussion? Take ownership of your content and moderation team, have some integrity and courage and stand behind your decisions.

Slashdot's censorship system is terrible. I see better discussions in youtube comments these days, believe it or not.

Thing about spam

[buss_error]

The thing about spam is that for as long as I can remember (at least back to 1997) people have insisted upon a technical solution for spam. The issue is that spam is not a technical problem. It's a human problem. Like any other problem/response cycle, if you are solving for the wrong issue, don't be shocked if the solution isn't as bad or worse than the problem. Another issue, not directly on point, is Google Email and anti-spam. I know of several organizations that have completely shut down their email infrastructure in favor of Google email services. An unaddressed problem is that these organizations have also laid off their email folks since "Google takes care of it all" so subtle and not so subtle issues often go not simply unaddressed, but unknown to the organization. The result has been a high rate of false positives, including senders without DKIM. I once got into a argument with John Lavine about DKIM, in which he got pretty passionate. I argue that DKIM is:

1. Needlessly opaque

2. Prone to abuse from over zealous admins

3. Google does it wrong (Checking the header chain all the way back instead of the last system the recipient does not run)

4. Breaks email standards

5. Doesn't solve any issue that SPF does not solve more directly, without possible abuse, and much more simply, requires far fewer CPU resources and skill, and does not break email standards in the process.
I'm told that "I'm too stupid" to know how it works and "I should get out of computers since you obviously are too stupid to know your f'ing job!" (both quotes from right here on slash dot). I won't try to prove otherwise, but one question I've asked over and over again is how DKIM, checked back further than the last untrusted relay, does not break email standards for list or forwarded mail. SPF won't break those, DKIM will, every time.

So getting back to our muttons, I'm not surprised that Google's spam engine (or anyone's, for that matter) has a high false positive rate, or a lower than desired true positive rate. That issue is simple - they are attempting to solve a problem with technology that isn't technical in nature. Stop using a hammer to try to screw in a light bulb. Doesn't work well.

Re:Time for a new search engine?

[Anonymous Coward]

My think my internal spam filter just got tripped.

Could you re-phrase that in coherent sentence/paragraph structure? Or was that a deliberate attempt at "conversational hypnosis" Illuminati mind control?

"Fake News" and "Spam" are now an excuse to censor

[sycodon]

I've seen it on Disqus too.

Google and Diqus and others provide tools to the site operators also who seem to use them to disappear comments without being accused of censoring.

I've had comments, all of the same political temperament and even the same text be "detected" as spam on some sites and not others.

Re:Time for a new search engine?

[AHuxley]

AC why try and filter the vast internet with all the comments about illegal migrants, blasphemy, news results about Tiananmen square and 1989?
Why not just create a safe space with an internet list? All the Hero Brigades SJW teams could add the few news sites they think are politically and culturally appropriate.

Focus on the ability to build a new internet. Why try and hold back all the sites that are not inclusive in real time?
Think of looking up authors or composers.
With a SJW list of approved arts sites the users would be only ever be presented with inclusive diversity.

No surprise

[grumbel5969]

How is that even a surprise? This is not a spam filtering tool that might catch a few wrong messages. This is explicitly advertised to perform censorship by getting rid of messages that might count as harassment or toxic. Since both of those are highly subjective, it will of course get rid of a lot valid messages, as that's it's job, that is what it was build for. If you want to automate your censorship, don't be surprise when your censorship happens automatically.

I find it ridiculous how much effort is spent on trying to cure toxicity, when most of it is the direct result of really basic usability flaws in the UI design. On Youtube for example you can only see the last 20 or so comments and higher ranked comments raise to the top. So of course you get clickbaity jokes and crap instead of good discussion, as you couldn't even have a good discussion within that comment system if you tried. Same with Twitter and it's 140 character limit. Comment system are more often than not so broken that you really can't hope to ever get good discussions out of them, no matter how much censorship you try.

Re:Time for a new search engine?

[AmiMoJo]

Google hasn't censored results for China since 2010. The Chinese government censors certain searches via their firewall because Google has results returned from servers in Hong Kong.